Advancing Innovative Data Processing Techniques by Leveraging AI and ML Technologies

Improving efficiency of identifying, tracking, and ageing fish from video and imagery
2023-present
Authors
Affiliations

S.Y. Alaba

Mississippi State University

J.E. Ball

Mississippi State University

R. Caillouet

NOAA Southeast Fisheries Science Center

M.D. Campbell

NOAA Southeast Fisheries Science Center

M.D. Grossi

NOAA Southeast Fisheries Science Center

R. Moorhead

Mississippi State University

M.M. Nabi

Mississippi State University

J. Prior

NOAA Southeast Fisheries Science Center

T.J. Rowell

NOAA Southeast Fisherise Science Center

C. Shah

Mississippi State University

F. Wallace

NOAA Southeast Fisheries Science Center

…and others

Screenshot of automatic species detection in action.

Screenshot of automatic species detection and classification in action, with bounding boxes showing fish found within the image and label indicating the most probable species name and uncertainty. Image courtesy of J. Prior.

Artificial intelligence (AI), machine learning (ML), and deep learning (DL) research and development has rapidly accelerated in recent years throughout NOAA. At the NOAA Fisheries Southeast Fisheries Science Center (SEFSC), innovative ML-based automations have been developed for ageing fish (e.g., red snapper, Atlantic menhaden), species identification, and tracking reef fish in underwater survey videos. Fish age is the fundamental parameter used for estimating growth, longevity, mortality, maturity, and reproductive output of specific species stock, while species counts provide estimates of abundance and diversity. All of this information gets feds into stock assessment models to enable science-based fisheries management. These automations have been developed by academic partners at the University of Washington and Mississippi State University in collaboration with SEFSC staff, with the goal of increasing the Center’s capacity to process data and the efficiency with which the Center is able to fulfill its mission of providing the scientific advice and data needed to effectively manage the living marine resources of the Southeast Region and Atlantic High Seas.

Papers and Conference Proceedings

  1. Alaba, S.Y., J.H. Prior, C. Shah, M.M. Nabi, J.E. Ball, R. Moorhead, M.D. Campbell, F. Wallace, and M.D. Grossi (2024) Multifish tracking for marine biodiversity monitoring, Proc. SPIE 13061, Ocean Sensing and Monitoring XVI, 130610E (6 June 2024), https://doi.org/10.1117/12.3013503.

  2. Shah, C., M.M. Nabi, S.Y. Alaba, R. Caillouet, J. Prior, M. Campbell, M.D. Grossi, F. Wallace, J.E. Ball, and R. Moorhead (2024) Active detection for fish species recognition in underwater environments, Proc. SPIE 13061, Ocean Sensing and Monitoring XVI, 130610D (6 June 2024), https://doi.org/10.1117/12.3013344.

  3. Shah, C., M.M. Nabi, S.Y. Alaba, I.A. Ebu, J. Prior, M.D. Campbell, R. Caillouet, M.D. Grossi, T. Rowell, F. Wallace, J.E. Ball, and R. Moorhead (2025) YOLOv8-TF: Transformer-Enhanced YOLOv8 for Underwater Fish Species Recognition with Class Imbalance Handling, Sensors, 25(6), 1846, https://doi.org/10.3390/s25061846.

  4. Shah, C., M.M. Nabi, I.A. Ebu, J. Prior, M.D. Grossi, F. Wallace, T. Rowell, J.E. Ball, R. Moorhead, R. Caillouet, M. Campbell (2025) Improved fish tracking in underwater images for marine biodiversity monitoring, Proc. SPIE 13460, Machine Learning from Challenging Data 2025, 134600F (29 May 2025), https://doi.org/10.1117/12.3053499.